recently updated kernel crashes during hammer

Pierre Abbat phma at leaf.dragonflybsd.org
Fri Jul 4 03:46:13 PDT 2014


Every day for the past several days, darner has been crashing.

Here's the top of top:

load averages:  0.28,  0.37,  0.20;               up 0+23:43:49        
06:11:47
32 processes: 1 running, 1 idle, 31 stopped
CPU states:  0.0% user,  0.0% nice, 65.8% system,  2.1% interrupt, 32.1% idle
Memory: 430M Active, 241M Inact, 268M Wired, 32M Cache, 110M Buf, 15M Free
Swap: 2048M Total, 1548K Used, 2046M Free
Timeout, server darner not responding.
phma at caracal:~/isp$ CE  SIZE    RES    STATE CPU  TIME   CTIME    CPU COMMAND
 62850 root         0  2796K   924K   hmrlck  1   0:00    0:00 43.95% hammer
   629 root         0  2776K   812K   nanslp  1   0:02    0:02  0.73% dntpd
 62428 root         0  3200K  1024K     wait  0   0:00    2:41  0.05% hammer

The screen shows this:

0xabf88490 does not look like a stack frame, skipping
0xabf88494 does not look like a stack frame, skipping
0xabf88498 does not look like a stack frame, skipping
Trace beginning at frame 0xabf8849c
(null)(abf884a0,0,ffffffff,0,0) at 0 0
(null)(0,abf884a0,0,ffffffff,0) at 0 0
_end() at 0xabf88498 0xabf88498
Debugger("panic")

CPU1 stopping CPUs: 0x00000001
 stopped
Stopped at      Debugger+0x3c: movb    $0,in_Debugger,4966
db>

Attempting to call dumpsys results in SECONDARY PANIC ON CPU 0 THREAD 
0xa0c80000. I can't get a dump.

This may have originated in a power blink which happened while I was asleep. 
I've seen it disconnecting ad1, or giving errors on the disk, as if there were 
not enough power to run the disk. But that didn't happen this morning, as far 
as I can tell.

$ uname -a
DragonFly darner.ixazon.lan 3.9-DEVELOPMENT DragonFly v3.9.0.125.ge809f-
DEVELOPMENT #0: Mon Jun 16 22:31:40 EST 2014     
root at darner.ixazon.lan:/usr/obj/usr/src/sys/GENERIC  i386

I installed world on June 18. How do I fix this?

Pierre
-- 
loi mintu se ckaji danlu cu jmaji




More information about the Users mailing list